Dealing with uncertainty has become a matter of great concern of policy makers and scientific research in a world facing global, epochal and complex changes. Such call for a comprehensive understanding of the implications of uncertainties lead to the broader question of: can we anticipate the future, and if so, how? This conceptual paper explores the ability to anticipate the future by converting undefined uncertainties into manageable risks. To do so, we distinguish between epistemic and ontological level of analysis, to argue that there are both epistemic and ontological limits in the substitu-tion of risk for uncertainty. We build in particular on the distinction drawn by Frank H. Knight (1921) between risk and true uncertainty to put forward such limits. Yet, in contrast to Knight who relies on expert judgment to overcome the epistemic limits, we suggest that the future cannot be fully anticipated through the production of new knowledge. Notwithstanding, we contend that we are better armed to face an unknown future with a co-production of knowledge by a larger range of actors. We illustrate our argument with the uncertainties arising from the current global ecological crisis.